Accurate ammonia spraying control method and system for low-temperature SCR denitration system

Abstract

The invention discloses a precise ammonia spraying control method and system for a low-temperature SCR denitration system, and belongs to the technical field of flue gas denitration control. The method comprises the following steps: collecting flue gas real-time operation data; calculating the absolute amount of NOx and the theoretical ammonia spraying amount; determining a baseline ammonia spraying flow based on the theoretical ammonia spraying amount, the ammonia water concentration and the evaporation efficiency; through a multi-dimensional dynamic prediction model fusing Kalman filtering and a neural network, the NOx concentration of an outlet is predicted, and a feedforward correction amount is generated; generating feedback correction according to the deviation between the actual concentration of the NOx at the outlet and a set value; and synthesizing the baseline flow, the feedforward correction and the feedback correction to obtain the final ammonia spraying flow. The system is further provided with a self-learning optimization module, the ammonia nitrogen molar ratio can be dynamically corrected, and the liquid-gas ratio is cooperatively adjusted according to the concentration of SO2 at an outlet of the desulfurization tower. According to the method, conversion of ammonia spraying control from passive response to active prediction is achieved, the denitration precision and the system stability are improved, desulfurization cooperation is considered, and the method is suitable for optimization and upgrading of the low-temperature SCR denitration system.

Description

Technical Field

[0001] This invention relates to the field of precision ammonia injection control technology for flue gas denitrification, specifically to a precision ammonia injection control method and system for low-temperature SCR denitrification systems. Background Technology

[0002] Nitrogen oxides (NOx) are among the main pollutants in flue gas from coal-fired power plants, sintering plants, and the steel industry. Emissions into the atmosphere cause environmental problems such as photochemical smog, acid rain, and ozone layer depletion, leading to increasingly stringent national and local emission standards. Currently, sintering machines typically employ SDA+SCR or CFB+SCR processes for desulfurization and denitrification. The SCR denitrification process uses 20% ammonia water as a reducing agent to reduce NOx in the sintering flue gas to N2+H2O under the action of a catalyst, achieving denitrification. Currently, ammonia injection systems evaporate ammonia water through an ammonia evaporator and then uniformly inject ammonia gas into the flue gas through an ammonia injection grid. Under the action of a catalyst, the ammonia reacts with NOx... x The reaction produces nitrogen and water, and the denitrification efficiency can reach over 90%, meeting the requirements for ultra-low emissions.

[0003] However, in actual operation, existing ammonia injection control methods still mainly rely on single-loop feedback control or manual adjustment, which leads to the following shortcomings in the denitrification system: NO in sintering flue gas x Frequent changes in concentration and flow rate, inlet NO x The system is significantly affected by load, temperature, and raw material fluctuations. Traditional single-loop feedback control suffers from response delays, which can easily lead to delayed or excessive ammonia injection, resulting in high NO levels at the outlet. x Large concentration fluctuations; manual adjustment is also difficult to respond to changes in operating conditions in a timely manner, often requiring operators to adjust the ammonia injection volume based on experience, resulting in poor system operation stability.

[0004] To improve the accuracy and response speed of ammonia injection control, some technological improvements have been made in related fields. For example: Patent CN119425377A discloses a nitrous oxide emission reduction control system and method for an SCR denitrification device. By monitoring and adjusting parameters such as flue gas temperature and ammonia injection flow rate, the reaction conditions are optimized to reduce nitrous oxide emissions. However, its control strategy does not involve short-term prediction of NOx concentration, nor does it adopt a dynamic feedforward mechanism, and it still belongs to passive response control.

[0005] Patent CN116651203A proposes an SCR denitrification control method that adapts to frequent fluctuations in unit load. It adjusts the ammonia injection rate by feeding forward multiple parameters such as inlet NOx concentration, SCR reaction temperature, and flue gas damper opening, and limits the ammonia injection rate when the temperature drop rate is high. However, it does not introduce a nonlinear dynamic prediction model, and its adaptability to the coupling effect of complex operating conditions is limited, so it cannot achieve true active intervention.

[0006] Patent CN117679930A relates to a precise ammonia injection method for flue gas coupled denitrification. It uses methods such as SNCR and SCR coupling and stratified ammonia injection to improve denitrification efficiency and control ammonia escape. However, its technical focus is on process structure optimization. It has not established a data-driven NOx concentration prediction model, nor has it achieved real-time feedforward correction of ammonia injection flow rate.

[0007] In summary, while existing technologies have made improvements in ammonia injection control, they still generally suffer from the following shortcomings: a lack of accurate short-term prediction capabilities for NOx concentration trends; a reliance on passive feedback control methods that cannot proactively adjust ammonia injection rates before changes in operating conditions; and the fact that existing methods often fail to consider the synergistic optimization of denitrification and desulfurization systems, leaving room for improvement in overall flue gas purification efficiency. Therefore, there is an urgent need for a precise ammonia injection control method capable of short-term NOx concentration prediction, feedforward feedback composite control, and self-learning and multi-system synergistic optimization capabilities to enhance the stability, economy, and environmental performance of low-temperature SCR denitrification systems. Summary of the Invention

[0008] To address the aforementioned technical problems, this invention proposes a precise ammonia injection control method and system for low-temperature SCR denitrification systems. By introducing intelligent technologies such as multi-dimensional dynamic prediction, feedforward-feedback composite control, self-learning optimization, and multi-system collaboration, the method can achieve real-time dynamic and precise adjustment of the ammonia injection quantity, demonstrating significant advantages in improving denitrification accuracy, stability, and economy.

[0009] This invention is achieved through the following technical solution: In a first aspect, the present invention provides a method for precise ammonia injection control in a low-temperature SCR denitrification system, comprising the following steps: Collect real-time operating data of the sintering flue gas system; Calculate the absolute amount of NOx based on the inlet NOx concentration and flue gas flow rate; The theoretical ammonia injection rate is calculated based on the absolute amount of NOx and the set ammonia-nitrogen molar ratio. The baseline ammonia injection flow rate is calculated based on the theoretical ammonia injection rate, ammonia concentration, and ammonia evaporator efficiency. A multidimensional dynamic prediction model that integrates Kalman filtering and neural networks is used to predict the future exit NO. x Concentration, generating ammonia injection feedforward correction amount; According to export NO x The deviation between the actual concentration and the set target value generates an ammonia injection feedback correction amount. By combining the baseline ammonia injection flow rate, the feedforward correction, and the feedback correction, the final ammonia injection flow rate is determined, thereby achieving precise ammonia injection control.

[0010] As a further optimization of the present invention, the real-time operating data also includes at least one of the following: flue gas temperature, sintering machine load, catalyst layer pressure difference, SCR outlet NOx concentration, outlet ammonia slip concentration, and desulfurization tower outlet SO2 concentration.

[0011] As a further optimization of the present invention, the formula for calculating the absolute amount of NOx is as follows: ; The formula for calculating the theoretical ammonia injection amount is: ; The formula for calculating the baseline ammonia injection flow rate is: ; in, Absolute amount of NOx in flue gas This refers to the concentration of ammonia. The efficiency of the ammonia evaporator.

[0012] As a further optimization of the present invention, the expression of the multidimensional dynamic prediction model is: ; in, To predict the future NOx concentration at constant outlet For Kalman filter prediction module, This is a nonlinear correction module for neural networks. For adaptive weighting coefficients, The inlet NOx concentration, For flue gas flow rate, T For flue gas temperature, L For the load of the sintering machine, For catalyst layer pressure difference, In the current Before that moment, The historical time series data of NOx concentration at the outlet is formed by backtracking n sampling times from the endpoint.

[0013] As a further optimization of the present invention, the calculation method of the ammonia injection feedforward correction is as follows: ; in, For ammonia injection feedforward correction amount This is the feedforward scaling factor. For the present Exit NO x Actual concentration.

[0014] As a further optimization of the present invention, the calculation method for the ammonia injection feedback correction amount is as follows: ; in, This is the ammonia injection feedback correction amount. For feedback ratio coefficient, Cset For export NO x Set a target value for the concentration. For the present Exit NO x Actual concentration.

[0015] As a further optimization of the present invention, the formula for determining the final ammonia injection flow rate is as follows: ; in, For a definite future The final ammonia injection flow rate at that moment, The baseline ammonia injection flow rate is [value missing]. This is the feedforward correction amount for ammonia injection. This is the ammonia injection feedback correction amount. This is the correction amount for ammonia slip.

[0016] As a further optimization of the present invention, a self-learning optimization step is also included: dynamically adjusting the ammonia-nitrogen molar ratio based on changes in denitrification efficiency, wherein the adjustment formula is: ; in, The corrected ammonia-nitrogen molar ratio. For the present ammonia-nitrogen molar ratio at any time For the present Denitrification efficiency at all times This refers to the learning rate.

[0017] As a further optimization of the present invention, a desulfurization synergistic control step is also included: dynamically adjusting the liquid-gas ratio based on the SO2 concentration deviation at the desulfurization tower outlet, wherein the adjustment formula is: ; in, The liquid-to-gas ratio, This is the desulfurization adjustment coefficient. Set a target value for the SO2 concentration at the outlet of the desulfurization tower. This refers to the SO2 concentration at the outlet of the desulfurization tower.

[0018] Secondly, the present invention provides a precision ammonia injection control system for a low-temperature SCR denitrification system, for implementing the aforementioned precision ammonia injection control method, comprising: The data acquisition module is used to collect real-time operating parameters of the flue gas denitrification system; The calculation module is used to calculate NO. x Absolute quantity, theoretical ammonia injection quantity, and baseline ammonia injection flow rate; The prediction module incorporates a multi-dimensional dynamic prediction model that combines Kalman filtering and neural networks to predict export NO. x Concentration and generate feedforward correction; Feedback adjustment module, used to adjust according to export NO x Concentration deviation generates feedback correction amount; The control module is used to integrate the baseline ammonia injection flow rate, feedforward correction, and feedback correction to output the final ammonia injection control command. An execution module is used to adjust the ammonia injection flow rate according to the control command.

[0019] Compared with existing technologies, the precise ammonia injection control method and system for low-temperature SCR denitrification systems proposed in this invention exhibit the following significant advantages and beneficial effects: 1. This invention uses a multi-dimensional dynamic prediction model, employing Kalman filtering for linear noise suppression and neural networks for nonlinear dynamic fitting, to achieve short-term trend prediction of NOx concentration. This enables the system to adjust ammonia injection before NOx concentration changes, thus transforming from a passive response to active control.

[0020] 2. This invention employs a baseline + feedforward + feedback composite control strategy. The baseline provides the basic ammonia injection rate, the feedforward responds quickly based on prediction results, and the feedback fine-tunes the output based on real-time deviations, forming a two-layer adaptive control mechanism. An ammonia slip correction mechanism is introduced; when excessive ammonia slip is detected, the ammonia injection rate is automatically reduced to prevent ammonium salt deposition on the catalyst surface, extend catalyst life, and improve the long-term stability and reliability of the system. The ammonia injection flow rate can be adjusted in real-time according to changes in flue gas conditions, avoiding excessive NOx emissions due to insufficient ammonia injection, or increased energy consumption and ammonia slip due to excessive ammonia injection.

[0021] 3. This invention achieves dynamic optimization of the L / G ratio by real-time feedback adjustment of the liquid-gas ratio based on real-time monitoring of the SO2 concentration at the desulfurization tower outlet. This enables the denitrification and desulfurization control strategies to be coordinated, further improving the overall flue gas purification efficiency.

[0022] 4. This invention incorporates a self-learning optimization module that dynamically adjusts the ammonia-nitrogen molar ratio based on changes in denitrification efficiency, automatically adapting to uncertainties during long-term operation such as catalyst activity decay and changes in operating conditions. The system can continuously optimize control parameters without human intervention, maintaining stable denitrification efficiency and reducing operating and maintenance costs.

[0023] 5. This invention adds only a data acquisition module, a prediction calculation module, and learning control logic to the existing control system, without changing the structure of the existing ammonia injection device, thus making it easy to promote and apply to existing sintering machine SCR systems. This method is suitable for low-temperature SCR systems of 280–350℃, and is particularly suitable for industrial scenarios with frequent operating condition fluctuations, such as sintering machines, demonstrating good engineering applicability and promising prospects for widespread application. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a block diagram of the control system of the present invention. Detailed Implementation

[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0027] The flowcharts used in this invention illustrate operations implemented according to some embodiments of the invention. It should be understood that the operations in the flowcharts may be implemented out of order, and steps without logical contextual relationships may be reversed or performed simultaneously. Those skilled in the art, guided by the content of this invention, may add one or more additional operations to the flowcharts, or remove one or more operations from the flowcharts. These functional entities may be implemented in software.

[0028] In this invention, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0029] This invention proposes a precise ammonia injection control method for a low-temperature SCR denitrification system, applied to a sintering machine flue gas denitrification system. The method mainly includes the following steps: Step 1: Real-time data acquisition: Collect real-time operating data of the sintering flue gas system, including inlet NOx concentration. flue gas flow rate Flue gas temperature T, sintering machine load L, catalyst layer pressure difference NOx concentration at SCR outlet ammonia escape concentration at the outlet and SO2 concentration at the desulfurization tower outlet ; In a real system, the above data can be collected by a variety of sensors placed in the flue gas duct, SCR inlet, upstream and downstream of the catalyst layer, and desulfurization tower outlet. For example, NOx concentration can be obtained through CEMS online monitoring device, flue gas flow rate is measured in real time by Pitot tube flow meter, and temperature, differential pressure and load parameters can be provided by temperature sensor, differential pressure sensor and PLC control system respectively. Then all signals are input into central control system through data acquisition module.

[0030] Step 2: Calculate the absolute amount of NOx: Based on inlet NOx concentration With flue gas flow To calculate the absolute amount of NOx in flue gas, use the following formula: ; This calculation step is used to quantify the pollutant load in the flue gas. Due to the frequent changes in sintering conditions, the NOx concentration and flow rate fluctuate greatly. If real-time calculation is not performed, the subsequent ammonia injection amount cannot reflect the true load, which can easily lead to fluctuations in denitrification efficiency.

[0031] Step 3: Calculate the theoretical ammonia injection rate: According to the preset ammonia-nitrogen molar ratio The theoretical ammonia injection rate is calculated using the following formula: ; in, The ammonia-nitrogen molar ratio under ideal chemical conditions is typically taken as 0.9 to 1.05.

[0032] The calculation yielded The theoretical mass flow rate of injected NH3, in mg / s, is used to ensure the stoichiometric reaction is adequate.

[0033] Step 4: Calculate the baseline ammonia injection flow rate: After obtaining the theoretical ammonia injection rate, the baseline ammonia injection flow rate is further calculated using the following formula: ; in: The ammonia injection flow rate is kg / s; The concentration of ammonia water is %; The efficiency of the ammonia evaporator is %.

[0034] The baseline ammonia injection rate represents the basic amount of ammonia that should be injected under the current load and operating conditions, providing an initial reference point for the entire control system.

[0035] Step 5: Determine the actual NOx concentration at the SCR outlet compared to the set target value. The deviation is addressed by continuously correcting the ammonia injection flow rate using a feedback control algorithm, thereby achieving dynamic and stable control of NOx emissions. Specifically, the control system continuously compares the actual detected values. With target value The difference between them is corrected by adjusting the opening of the ammonia injection valve through feedback, so that the outlet NOx is stabilized within the target range.

[0036] Furthermore, in a preferred embodiment of the present invention, the ammonia injection control adopts a composite control strategy of baseline + feedforward + feedback. Specifically, the control of the ammonia injection flow rate includes: Based on the ammonia injection feedforward correction amount of the prediction model, respond in advance to changes in system operating conditions; The output is dynamically fine-tuned based on the ammonia injection feedback correction amount based on the real-time concentration deviation. The two work together, with the feedforward link providing rapid prediction and the feedback link providing a stable closed loop, thus forming a two-layer adaptive control mechanism.

[0037] Before feedback control, a multidimensional dynamic prediction model based on Kalman filtering (KF) and neural networks (NN) is established to predict the outlet NOx concentration at future times. The model expression is as follows: ; in: This is a Kalman filter prediction module used to remove noise and capture linear trends. This is a neural network nonlinear correction module used to identify complex coupling effects during sudden changes in operating conditions. The weighting coefficients are adaptive and dynamically updated based on the prediction residuals; the prediction results are used to correct the ammonia injection flow rate in advance.

[0038] The introduction of Kalman filtering can effectively eliminate the interference of sensor noise on the prediction results, while neural networks can compensate for nonlinear errors, thereby achieving a fusion prediction of linear and nonlinear results.

[0039] The prediction results are used to calculate the ammonia injection feedforward correction. Based on the prediction results, the ammonia injection feedforward correction is calculated. Calculate according to the following formula: ; in, This is the feedforward scaling factor; If an increase in NOx emissions is predicted, the system increases the ammonia injection flow rate in advance; conversely, it reduces the ammonia injection flow rate in advance if the increase is not predicted. Through this mechanism, ammonia injection control can shift from a passive response to an active adjustment, thereby significantly shortening the system response time.

[0040] Furthermore, in a preferred embodiment of the present invention, it further includes ammonia injection feedback correction: calculating the feedback correction amount based on the deviation between the actual outlet NOx concentration and the set target value. Specifically, it is calculated using the following formula: ; in, This is the feedback ratio coefficient; When the NOx concentration at the outlet is higher than the set value, the ammonia injection rate is increased; conversely, the ammonia injection rate is decreased. The feedback loop adjusts the ammonia injection rate in real time based on the outlet concentration error, thereby enabling the system to quickly return to a stable state after a disturbance.

[0041] The calculation of the overall ammonia injection rate involves the control system comprehensively calculating the final ammonia injection output based on the baseline ammonia injection rate, feedforward correction, feedback correction, and ammonia slip correction. The specific formula is: ; in, This is the correction amount for ammonia slip. When the outlet ammonia escape concentration is detected When the value is greater than 1 ppm, a negative correction is generated. <0 automatically reduces ammonia injection flow rate to prevent ammonium sulfate deposition and activity reduction on catalyst surface; The ammonia injection actuator automatically adjusts the proportional valve opening based on the calculated value, thereby achieving precise ammonia injection control.

[0042] Furthermore, in a preferred embodiment of the present invention, in order to maintain the system in optimal condition over a long period of time, the present invention also includes a self-learning optimization module, which optimizes the denitrification efficiency. The ammonia-nitrogen ratio is automatically corrected for changes in the ammonia-nitrogen ratio. The specific calculation formula is as follows: ; in, For learning rate, Limited to the range of 0.9 to 1.1; when The ammonia nitrogen ratio continued to decrease, and the system automatically and slightly increased it. To compensate for the catalytic activity decay; when If the temperature is too high and ammonia escape increases, it will automatically decrease. To prevent excessive ammonia spraying; Therefore, the algorithm can continuously optimize parameters and adjust ammonia injection strategy without human intervention, thereby adapting to catalyst aging and changes in operating conditions and ensuring stable denitrification efficiency.

[0043] Furthermore, considering the coupling relationship between the desulfurization system and SCR denitrification, this invention also automatically adjusts the liquid-to-gas ratio (L / G) by monitoring the SO2 concentration deviation at the desulfurization tower outlet. ; in, The desulfurization adjustment coefficient is used, and the L / G adjustment rate is limited to 8% / min. When the SO2 concentration is too high, the L / G ratio is increased to enhance the absorption capacity; when the SO2 concentration is too low, the absorbent flow rate is reduced to save energy.

[0044] Verification has shown that the ammonia injection control method provided by this invention can achieve a stable outlet NOx concentration of 40–50 mg / m³ under operating conditions of 280–350°C. 3 Denitrification efficiency ≥90%; ammonia slip ≤1ppm; SO2 emission ≤35mg / m³ 3 In addition, the fluctuation of the ammonia injection system was reduced by about 40%, the ammonium salt deposition on the catalyst surface was significantly reduced, and the system's operational stability was also significantly improved.

[0045] In summary, this invention achieves intelligent and precise ammonia injection control in low-temperature SCR systems by establishing a multi-dimensional prediction model, introducing feedforward and feedback dual-layer regulation, and combining ammonia slip correction and self-learning optimization. This method can be directly upgraded and implemented on existing devices without large-scale modifications, and can maintain stable denitrification efficiency, low energy consumption, and low pollutant emissions for a long time, thus having broad prospects for promotion and application.

[0046] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0047] The contents not described in detail in this specification are existing technologies known to those skilled in the art.

Claims

1. A method for precise ammonia injection control in a low-temperature SCR denitrification system, characterized in that, Includes the following steps: Collect real-time operating data of the sintering flue gas system; Calculate the absolute amount of NOx based on the inlet NOx concentration and flue gas flow rate; The theoretical ammonia injection rate is calculated based on the absolute amount of NOx and the set ammonia-nitrogen molar ratio. The baseline ammonia injection flow rate is calculated based on the theoretical ammonia injection rate, ammonia concentration, and ammonia evaporator efficiency. A multidimensional dynamic prediction model that integrates Kalman filtering and neural networks is used to predict the future exit NO. x Concentration, generating ammonia injection feedforward correction amount; According to export NO x The deviation between the actual concentration and the set target value generates an ammonia injection feedback correction amount. By combining the baseline ammonia injection flow rate, the feedforward correction, and the feedback correction, the final ammonia injection flow rate is determined, thereby achieving precise ammonia injection control.

2. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, The real-time operating data also includes at least one of the following: flue gas temperature, sintering machine load, catalyst layer pressure difference, SCR outlet NOx concentration, outlet ammonia escape concentration, and desulfurization tower outlet SO2 concentration.

3. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, The formula for calculating the absolute amount of NOx is: ； The formula for calculating the theoretical ammonia injection amount is: ； The formula for calculating the baseline ammonia injection flow rate is: ； in, Absolute amount of NOx in flue gas This refers to the concentration of ammonia. The efficiency of the ammonia evaporator.

4. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, The expression for the multidimensional dynamic prediction model is: ； in, To predict the future NOx concentration at constant outlet For Kalman filter prediction module, This is a nonlinear correction module for neural networks. For adaptive weighting coefficients, The inlet NOx concentration, For flue gas flow rate, T For flue gas temperature, L For the load of the sintering machine, For catalyst layer pressure difference, In the current Before that moment, The historical time series data of NOx concentration at the outlet is formed by backtracking n sampling times from the endpoint.

5. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 4, characterized in that, The calculation method for the ammonia injection feedforward correction is as follows: ； in, For ammonia injection feedforward correction amount This is the feedforward scaling factor. For the present Exit NO x Actual concentration.

6. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, The calculation method for the ammonia injection feedback correction amount is as follows: ； in, This is the ammonia injection feedback correction amount. For feedback ratio coefficient, Cset For export NO x Set a target value for concentration. For the present Exit NO x Actual concentration.

7. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, The formula for determining the final ammonia injection flow rate is as follows: ； in, For a definite future The final ammonia injection flow rate at that moment, The baseline ammonia injection flow rate is [value missing]. This is the feedforward correction amount for ammonia injection. This is the ammonia injection feedback correction amount. This is the correction amount for ammonia slip.

8. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, It also includes a self-learning optimization step: dynamically adjusting the ammonia-nitrogen molar ratio based on changes in denitrification efficiency, using the following formula: ； in, This is the corrected ammonia-nitrogen molar ratio. For the present ammonia-nitrogen molar ratio at any time For the present Denitrification efficiency at all times This refers to the learning rate.

9. The precise ammonia injection control method for a low-temperature SCR denitrification system according to claim 1, characterized in that, It also includes a desulfurization synergistic control step: dynamically adjusting the liquid-gas ratio based on the SO2 concentration deviation at the desulfurization tower outlet, with the following adjustment formula: ； in, The liquid-to-gas ratio, This is the desulfurization adjustment coefficient. Set a target value for the SO2 concentration at the outlet of the desulfurization tower. This refers to the SO2 concentration at the outlet of the desulfurization tower.

10. A precision ammonia injection control system for a low-temperature SCR denitrification system, used to implement the precision ammonia injection control method according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to collect real-time operating parameters of the flue gas denitrification system; The calculation module is used to calculate NO. x Absolute quantity, theoretical ammonia injection quantity, and baseline ammonia injection flow rate; The prediction module incorporates a multi-dimensional dynamic prediction model that combines Kalman filtering and neural networks to predict export NO. x Concentration and generate feedforward correction; Feedback adjustment module, used to adjust according to the export NO x Concentration deviation generates feedback correction amount; The control module is used to integrate the baseline ammonia injection flow rate, feedforward correction, and feedback correction to output the final ammonia injection control command. An execution module is used to adjust the ammonia injection flow rate according to the control command.

Images